On Feb 10, 2012, at 2:05 PM, Ivica Ico Bukvic wrote:
Why have two wiimote objects? Is there a specific reason why they can't be merged? If we pool the work on it we'll have one better wiimote object versus two different wiimote objects that are both less good.
Do you mind elaborating why you believe both objects are "less good" in and of themselves? IMO disis_wiimote is much more robust than the wiimote object (at least the last time I tested) in every respect and provides entirely xrun-free functionality. If we agree the aforesaid statement is true (not that I am suggesting that you do agree) then it is not that both are less good but rather the one is better than the other literally in every respect which should in theory make the whole process of selection a lot simpler.
Also, suggested pooling may not be always the best approach, particularly if there is limited transparency in terms of how such process is managed (which BTW I think is the core reason why pd-l2ork and disis_wiimote exist in the first place, and consequently why pd community at large is having trouble with the uptake). Also, let us not forget that at the very core of the open-source idea is the "survival of the fittest" model (with a caveat, however; read * below). This means if one external works better than another, it will simply supersede its operation, particularly if their output is essentially the same. Renaming it (or using other practical alternatives to "pooling"), provided credit is given where credit is due, is at this point irrelevant and should be considered synonymous to pooling, particularly in this case where disis_wiimote AFAIK does everything wiimote does except it does it in a way that is (IMO) more robust**
*Of course, any dev has every right to continue to maintain their own version for whatever reason and thus defy the context of the survival of the fittest regardless whether they are maintaining one of the "fittest" or less "fit" versions. Also, obviously everyone is entitled to their own definition of what constitutes "fittest" iteration...
**Based on 2.5+ years of experience of having a bunch of students with no prior Linux/PD experience messing with the system, including the wiimote external. I hope others would agree that the object's stability is best tested when used by others than the developer who usually focuses on aspects that best cater to their own needs and that may not be always as all-encompassing to cover all use cases. Thus, the greatest collection of instabilities will surface through third-party use. The same is true for pd as a whole...
Best wishes,
For one thing, I cannot "apt-get install pd-disis-wiimote". Another thing, I know that pd-wiimote has had quite a bit of testing when using 6 wiimotes simultaneously.
You are of course free to do whatever you want with the code, no one is challenging that. I am talking about contributing patches upstream. That's widely accepted as the best practice, and then forking is a good second option.
.hc
Man has survived hitherto because he was too ignorant to know how to realize his wishes. Now that he can realize them, he must either change them, or perish. -William Carlos Williams